Is there a litmus test to know whether something is science or pseudoscience? There are many things which is quite ambiguous like ayurveda, homeopathy, psychology, biology, etc...

  • 42
    Homeopathy is not ambiguous, it's pseudoscience pure and simple.
    – armand
    Commented Aug 15, 2022 at 21:20
  • 11
    @armand: Ayurvedic medicine is not ambiguous either. Why single out homeopathy when essentially every kind of 'traditional medicine' is full of pseudoscience?
    – user21820
    Commented Aug 16, 2022 at 12:36
  • 10
    @user21820 because I don't know much about ayurvedic stuff and I try to avoid talking about what I don't know. Homeopathy I know, and i see it as the epitome of pseudoscience, as it lacks any experimental or theoretical ground yet manages to pass as science through cargo cult mimicry (labs, lab coats, test tubes, etc...)
    – armand
    Commented Aug 16, 2022 at 13:03
  • 7
    @JimmyJames let's never forget the saying: "how do we call alternative medicine that work? Well, we call it medicine."
    – armand
    Commented Aug 16, 2022 at 16:08
  • 11
    Regarding "There are many things which is quite ambiguous like ayurveda, homeopathy, psychology, biology, etc...": How did Biology make it into that list?
    – Nat
    Commented Aug 17, 2022 at 3:29

16 Answers 16


This question is closely related to a problem in philosophy of science known as the demarcation problem. A good starting point for broad overview would be the SEP article.

To answer you question directly, I first need to clarify what you mean by "litmus test". I will assume you mean that there is a short checklist of criteria that can be fairly easily evaluated such that whenever something which purports to be science meets enough of those criteria we can be confident it really is science. If it does not meet enough of the criteria then we can determine it is pseudoscience.

If something roughly like this is what we mean by "litmus test" then the answer is clear no. There is no such simple test that distinguishes science from pseudoscience in every case.

The most common thought for how to construct a test would be to say that real science is research conducted in accordance with the scientific method, while pseudoscience does not. However, this just turns the original problem into a problem of how to operationalize "the scientific method". Nobody has has solved this in way that is uncontroversial, universal, and sufficiently specific. There are proposals which are detailed and specific, and which might thus be used as some kind of effective test. However, these have limited scope (only covering certain types of scientific disciplines or only pertaining to particular stages in the historical development of the sciences) and they usually represent a specific viewpoint within the scope of legitimate scientific disagreement.

On the other hand, often people invoke broad generic characterizations of science such as "Science is systematically gathering evidence in order to construct and test abstract theories". This might be sufficient to rule out certain activities from being science, for example, when somebody who just invents a theory and never tries to test it with any sort of evidence but nevertheless describes what they are doing as "science". Such cases might be called pseudoscience, and could easily be rejected as not being science based on not seeking or responding to evidence (so in this sense there is a limited litmus test). However, it could be misleading to call such examples pseudoscience since they don't even have the appearance of science. Generally, to be pseudo-something, you need to appear to be the real deal while actually being ersatz. We might be better to categorize such claims to being "science" as examples of ideology, hucksterism, cultish thinking, or conspiracy theory.

It is only when people put real effort into at least appearing that they are gathering evidence and making arguments for their theories that questions of science vs. pseudoscience raise distinctive issues. For example, think of parapsychology researchers or climate change denying climatologists or anti-vaccine doctors. (I do not mean to imply that those three examples all have equal epistemic status, just that all might be considered pseudoscience by some.) In cases like these (which are the sort of cases that matter) no generic account of scientific method will be sufficient to differentiate science and from pseudoscience.

Another issue is that even within the legitimate sciences and legitimate scientific research institutions, there are many examples of research that is sloppy, biased, or even fraudulent. It would not be right to mistake badly done science with pseudoscience (although if an entire field were riddled with such poor methodology we might say this). So any proposed litmus test would need a way not just to evaluate specific research as good or bad science but to determine whether this research is part of a larger discipline of scientific research that is somehow self-correcting. Legitimate sciences respond to fraud and scandal by trying to develop mechanisms to prevent similar abuses in the future. However, these protective mechanisms are never perfect and change over time.

Finally, these considerations point to something else. There is no all-or-nothing, once-and-for-all answer to the question of whether something science or pseudoscience. It is a matter of degree how epistemically reliable a given discipline's scientific methodologies are and this reliability fluctuates over time.

In short, we should not be seeking a simple test for whether something is science or pseudoscience. Instead we should ask for an understanding of what makes research programs more or less based on reason and evidence, as well as more or less self-correcting in the face of abuses, corruption, and ideology. Such an understanding would be quite complex and not amenable to being turned into a simple test. This is the kind of knowledge one develops from through a broad education in critical thinking and by studying the history and philosophy of science.

  • 6
    "This is the kind of knowledge one develops from through a broad education in critical thinking and by studying the history and philosophy of science." arguably should be mandatory for every citizen of a democracy because they will need to vote. Commented Aug 16, 2022 at 5:10
  • Just as a note: Sometimes the standards of rigor go too far. Here for medicine it is very important to be evidence based (which scientist here use as a separation from pseudo medicice, lets call it). In the beginning of Corona there was the discussion that there is no scientific evidence that masks help. Head of medical association said things to this effect in nov 2020. So sometimes I would like to add the principle 'extraordinary claims need extraordinaire evidence' and 'resonable claims on current understanding sometimes dont need evidence (in the short run)'.
    – lalala
    Commented Aug 16, 2022 at 22:00
  • 1
    @lalala: the contrast terms in medicine I'm familiar with are evidence based vs. eminence based. However, I'd say another test that helps is to have a look how (un)certainty about is communicated. "Masks help" vs. "We think it plausible that masks help". "We don't have evidence" was likely true and a statement carrying information. It would have been even far more informative with an additional statement on what attempts to find evidence had been made. I fully agree that some claims are of so little weight or importance that they are not worth spending the ressources to study them.... Commented Aug 17, 2022 at 10:14
  • 1
    This answer reminds me of: xkcd.com/435. also this will be a great read to the question askwer: en.wikipedia.org/wiki/Falsifiability
    – Hakaishin
    Commented Aug 17, 2022 at 14:00
  • 1
    ... However, wrt. to topics where the scientific claim influences e.g. politics, its effect should also be considered when deciding how much certainty is required. Personally, a plausible claim that masks help against the spread of Covid is sufficient for me to openly discuss e.g. wearing a mask before visiting "grandma", I also accept it as basis for a general recommendation, but when it turns into regulation, I'd like to see that at least the regulation is accompanied by a well-designed scientific study and a corresponding revision date for that regulation. Otherwise we may end up in... Commented Aug 17, 2022 at 16:30

While you already got the very long (and good) answers showing you all the problems with the term "pseudo science" from a deep philosophical viewpoint, there are "litmus tests" that we as lay persons (in any given area) can apply to get a pretty quick feeling about whether something that purports to be science should influence our individual thinking and acting.

  • Is a given claim falsifiable? If I claim something, i.e., if I put out a theory of X, and there is a clearly defined way to show that my theory is wrong, then this is a good sign. This gives other people (or myself) the chance to keep working on the theory, and test it against the real world. The more people try to falsify it, and the more they fail, the more reliable the theory is. I'd say that this is the core of what would most people would call "science".
  • Does it make interesting predictions which cannot be experimentally verified right now (maybe because we don't have the technology yet)? These are especially good - they are bold claims, and when they eventually do get experimentally confirmed, lead to a quite high trust in the theory.
  • Is a theory evolving, or just appearing at once out of the blue moon in a finished form? The first is a good sign. It means that many people have worked on it, and the more people work on it, the more likely it is that people find errors and lead the work into a better direction.

These aspects are just some which are discussed in the context of the Scientific Method, and it is all a gray area (especially maybe the third point). As you see there, it is indeed a very complex field. The Wikipedia article on pseudo-science has lots and lots of markers that can be used to test this.

Another more indirect aspect which is maybe not strictly philosophical in nature, but increasingly important today would be:

  • Who benefits if the general public, and especially legislatives or government agencies accept the theory? Could there be ulterior motives? Some modern-day cases are pretty clear and do not require any cynicism or conspiracy theories, but just a general understanding of how funding, marketing etc. work in our global economy.
  • (From user usul in the comments): Is the theory attempting to be systematic and general? Good science attempts to discover underlying mechanisms or causes that are universally true and has implications in many scenarios. Be suspicious of claims like 'this particular object has this particular effect in this particular circumstance' or 'this particular statistic suggests this particular trend is true' if there is no attempt to connect to a more general mechanism of "how" or "why" that is relevant beyond the current scope.
  • 2
    These are good epistemic litmus tests. I would add: "is the theory attempting to be systematic and general?" Good science attempts to discover underlying mechanisms or causes that are universally true and has implications in many scenarios. Be suspicious of claims like 'this particular object has this particular effect in this particular circumstance' or 'this particular statistic suggests this particular trend is true' if there is no attempt to connect to a more general mechanism of "how" or "why" that is relevant beyond the current scope.
    – usul
    Commented Aug 16, 2022 at 15:29
  • Took it verbatim, @usul, thanks.
    – AnoE
    Commented Aug 17, 2022 at 7:50
  • While the proposed checks can be useful, they also fail for non-trivial portions of well-respected sciences. The tests are aimed at sciences which are theory-heavy, whereas some fields do not really build theories (e.g., descriptive biology) or make predictions (e.g., most of paleontology). You can stretch the definitions of "theory", "prediction", ... to include those cases, but I don't think it is very useful. On the other hand both astrology and conspiracy theories are falsifiable, make predictions and evolve over time. Honest critical thought about a particular case is thus always needed. Commented Aug 18, 2022 at 9:12

From the Stanford Encyclopedia of Philosophy's article on Pseudoscience:

There is widespread agreement for instance that creationism, astrology, homeopathy, Kirlian photography, dowsing, ufology, ancient astronaut theory, Holocaust denialism, Velikovskian catastrophism, and climate change denialism are pseudosciences. (Emphasis mine)

The fact that the US court system and others routinely have to deal with the reality of non-scientific practice being asserted as science is just an inescapable aspect of life, sorry. (See Kitzmiller) While many eggheads aim for deductive certainty before declaring something absolutely, positively objectively true, one can apply other standards of truth, such as pragmatic theories of truth. The word 'pseudoscience' is used widely, and with good purpose. There's A LOT of people claiming to do science that are doing NOTHING of the sort. How do philosophers characterize pseudoscience? Well, reading the article in full is your best path to knowledge, but here are a few major points from the article:

An Indifference to Epistemological and Empirical Foundations

Read section 6.4.

The term “bullshit” was introduced into philosophy by Harry Frankfurt, who first discussed it in a 1986 essay (Raritan Quarterly Review) and developed the discussion into a book (2005). Frankfurt used the term to describe a type of falsehood that does not amount to lying. A person who lies deliberately chooses not to tell the truth, whereas a person who utters bullshit is not interested in whether what (s)he says is true or false, only in its suitability for his or her purpose. Moberger (2020) has proposed that pseudoscience should be seen as a special case of bullshit, understood as “a culpable lack of epistemic conscientiousness”.

Many people who make pseudoscientific claims are well-educated, well-intentioned, but simply lack epistemic capacity and conscientiousness. A perfect example of such real-world tom foolery is anyone trying to pass-off multiverse pseudoscience (Hossenfelder) or the simulation hypothesis (more pseudoscience) (Hossenfelder) as science. According to SEP's Epistemology of Modality's section on conceivability, the notion of conceivability:

The main idea of (E2-D) is that there are two different ways in which we can evaluate statements across possible worlds, i.e., two different ways of conceiving hypothetical situations, based on two different constraints. The first constraint binds what is true in some possible world to what one knows a priori. A statement is primarily conceivable if nothing that is knowable a priori is incompatible with the statement being true... By contrast, it has been discovered empirically that water is H2O.

Some simply cannot understand or refuse to accept that empirical necessity is an unavoidable part of fallibilism (IEP), one which clearly has a role for modal empiricism and empiricism and physicalism more broadly. Watering down empirical positions and attacking empirical strawmen accompanies protests from pseudoscientists.

The Demarcation Problem Is Not the Denial of Dichotomy

There are things that are clearly not scientific being passed off as such is clearly a consensus of philosophers of science. Again, from SEP:

Kuhn observed that although his own and Popper’s criteria of demarcation are profoundly different, they lead to essentially the same conclusions on what should be counted as science respectively pseudoscience (Kuhn 1974, 803). This convergence of theoretically divergent demarcation criteria is a quite general phenomenon. Philosophers and other theoreticians of science differ widely in their views on what science is. Nevertheless, there is virtual unanimity in the community of knowledge disciplines on most particular issues of demarcation.

Some Criteria

The three big ones from the logical positivists are falsification (IEP), confirmation, and verification. None of them are perfect, but they are good rules of thumb.

The SEP even in section 4.6 offers an example list:

  • Belief in authority: It is contended that some person or persons have a special ability to determine what is true or false. Others have to accept their judgments.
  • Unrepeatable experiments: Reliance is put on experiments that cannot be repeated by others with the same outcome.
  • Handpicked examples: Handpicked examples are used although they are not representative of the general category that the investigation refers to.
  • Unwillingness to test: A theory is not tested although it is possible to test it.
  • Disregard of refuting information: Observations or experiments that conflict with a theory are neglected.
  • Built-in subterfuge: The testing of a theory is so arranged that the theory can only be confirmed, never disconfirmed, by the outcome.
  • Explanations are abandoned without replacement. Tenable explanations are given up without being replaced, so that the new theory leaves much more unexplained than the previous one.


There's no universal tool for categorizing pseudoscience, and certainly there are cases that strain the thinker, but pseudoscience is a frequently used rhetorical tool to try to smuggle clearly non-empirical, non-scientific thought into forums that teach science; sometimes those that protest the classification pseudoscience the loudest are those pushing pseudoscience claims.

  • 2
    Of the pseudosciences SEP listed, many of them do not claim to be science, so they aren't pseudoscience. And "climate change denialism" is either plain skepticism of experts who are often wrong, or scientific criticism of the media narrative. But those who go against the politically-approved science have always been attacked. It's no different than it was in the day of Galileo. Commented Aug 16, 2022 at 5:18
  • I like this! If I may add to the "Some Criteria" section, one thing I have found useful is to look at the different actors who are capable of testing something. There is science/pseudoscience which a layman can test in their home, and there is science/pseudoscience which requires multi-billion dollar facilities to test. I find that sliding scale helps with the demarcation problem, explaining why some quantum/cosmological theories get to be considered "science" while being untested (its just mind-numbingly expensive to test them), while other theories do not.
    – Cort Ammon
    Commented Aug 17, 2022 at 15:27
  • @DavidGudeman Fascinatingly, the climate-change skeptics are the ones who argue that the climate-change theories are pseudoscience, using many of the arguments listed here. If anything, I think that shows the tremendous challenge of the demarcation problem between science and pseudoscience. In the really hard problems, it takes a lot of skill to draw a meaningful line between the two.
    – Cort Ammon
    Commented Aug 17, 2022 at 15:28
  • @CortAmmon, I've not read extensively from the climate-change skeptics, but I don't recall them accusing climate-change advocates of engaging in pseudo-science; they just had scientific disagreements with them--disagreements on methodology, extrapolations, statics, etc. What they say about the journals that suppress skeptical papers, the general press, and the governmental bodies is a bit harsher. Commented Aug 17, 2022 at 15:35
  • @CortAmmon I think it's fair to consider that genuine denialism is a psychological phenomenon more than a philosophical one, and that counter-accusation is a form of projection. In a clash of worldviews, what is at stake is too emotionally charged not to affect the normative presumptions which I believe to be determined abductively to reach the conclusion to support the worldview. The interplay between the logic and emotion is too tightly coupled not to see the process of denialism not only in an epistemic light, but also in terms of emotional needs; thus, first principles are a bellwether.
    – J D
    Commented Aug 17, 2022 at 15:49

Science is to pseudoscience as truth is to bullshit. The key feature of pseduoscience is not whether it is scientific. It's that it doesn't care.

Just as bullshit might happen to contain truth or lies, pseduoscience might happen to contain elements of rigorous science. (Usually not all of them, because science is hard.) But the key point is that it is not genuine: it does not attempt to follow scientific principles and advance human knowledge and understanding, except by accident or as useful to accomplish its external goal.

This makes a litmus test difficult. A bullshitter may find that saying true things helps accomplishes her goal, but what she says cannot be relied on. Similarly, pseudoscience may incorporate actual scientific principles and practices to accomplish its goal. But it is willing to abandon them when convenient. So the results it produces cannot be relied on.

My stance is similar to Moberger (2020) as cited by JD, but I differ in arguing that pseudoscience is analogous to bullshit rather than a special case.

Pseudoscience is related to cargo cult science, but the latter often represents a genuine, but misguided, attempt to do science.



We bin aspects of our world into categories, but our binning always has fuzzy edges. You will not find a bombproof definition of science vs pseudoscience.

A little elaboration

Pragmatically, however, we find it very useful to distinguish one bin from another. And figuring out what science is from pseudoscience is VERY useful for figuring out who an expert is in a subject. the science/pseudoscience boundary is therefore one we will be well served to find a pretty good approximation for.

Karl Popper's criteria, that something is only science if it is falsifiable, is an excellent shorthand summary of this boundary.

More complexity

Philosophers are very good at critiquing each other, and Popper's criteria came in for a lot of criticism. One is that falsificationism is not really absolute. Actual scientists still hold by views even when they have falsifications of them. For example, for decades, the Big Bang theory said the universe was younger than our more reliable stellar ages model said the oldest stars in the Milky Way were. This refutation was only solved a few years ago when we discovered that our Cosmological Constant is changing. The Big Bang model was still accepted and considered true, despite the "refutation". A second is that logically refutations are impossible, as one can always kluge a theory to account for any inconvenient fact.

Another philosopher (Imre Lakatos) came up with a better description of science, as "Research Programmes" which involve a particular approach and assumption set, but which can have a whole family of related hypotheses that can be called that "theory". The Research Programme can at any time have up to dozens of "refutaitons" but it remains useful if it solves other problems, AND the researchers admit to and are working to try to solve the refuting "problems".

Note the importance of actually doing the tests that identify the problems, admitting to them, then trying to address them. This is an ATTITUDE of seeking and then addressing falsifications. If a Research Programme does this, and does peer review, peer critique, improvement in methodology, and follows good experimental/investigatory/statistical/mathematics practices, then such a research programme is science.

Misuse of the Term

There is a current movement of Organized Skeptics that misuse and throw the term Pseudoscience around as a rhetorical ploy, labeling anything that does not assume reductive materialism to be unquestionable, as "pseudoscience". This movement, which itself denies the possibility of the refutation of reductive materialism, is committed to rejecting falsifiability of its premises in any way. IE, this movement is anti-science, and by claiming to be "scientific", it is the prototype of a pseudoscience movement. You will find most accusations of "pseudoscience" in popular literature are from members of this pseudoscience movement.

Side issues

As I noted at the outset, concepts often have fuzzy boundaries. Karl Popper was very aware of this relative to science. He did not consider it to be easy, or useful, to try to distinguish philosophy from science. This is because sciences are emergent from, and refinement of, a particular philosophic project which proved highly useful in spawning a new specialty study area.

He also did not consider it easy or useful to distinguish sciences from informal empiricism, such as how people learned how to cook, or fletch an arrow. Science has just borrowed the empirical process, and tried to formalize how to apply it.

Sorting science from pseudoscience Popper considered to be socially far more important, so his criteria is only really useful to try to define that edge of science vs. non-science.

  • +1 for explaining the historical Popper had a lot more nuance than many of his modern critics readily admit.
    – J.G.
    Commented Aug 17, 2022 at 14:51

Science involves demonstrable claims, concrete scientific studies with statistically significant results, going with the evidence and not making unwarranted assumptions.

Pseudoscience involves claims based on studies with results that don't support the underlying claims, studies with severe methodological problems, denying or cherry-picking evidence and/or assuming things without evidence.

There is nothing ambiguous about homeopathy, psychology or biology, and it's curious that you'd group them together.

Homeopathy is clearly and always pseudoscience, while psychology and biology are clearly science. There may be pseudoscientific claims within the fields of psychology and biology, but unlike homeopathy, the fields as a whole are not pseudoscience, and there are plenty of non-pseudoscientific claims within those fields.

Psychology is often considered more of a "soft science", but I haven't heard the same prominently said about biology (except potentially by e.g. transphobes and gender critical people who want to dismiss the well-supported scientific fact that biological sex doesn't have a clear binary distinction). Although the distinction between "soft" and "hard" science is often used to dismiss legitimate scientific statistically-significant results. Although this distinction is getting a bit beyond the scope of this question and answer.

I don't know much about Ayurveda, but Wikipedia tells me it's alternative medicine that uses herbal remedies, etc. There's a joke I've heard a few times: "You know what you call alternative medicine that works? Medicine." If an alternative medicine treatment can be shown to work in a way that meets scientific standards, it would no longer be considered alternative medicine: it would just be regular medicine. This is especially relevant to herbal remedies (where demonstrable remedies are generally converted to regular medicine by turning them into pills). Although things like meditation and yoga (also included in Ayurveda) may have concrete and demonstrable health benefits (but any claims of such benefits should also meet scientific standards to classify as scientific instead of pseudoscientific).


There is indeed a simple litmus test: The presence/absence of scientific rigor.

Any science that is worth its salt will, within the boundaries of its subject

  • measure facts

  • build theories

  • test theories with experiments

  • discard theories that fail the test of the experiment.

This is the only way that we can build up a pool of knowledge that we can trust.

Note that the terms in the above are rather broad: A measured fact can be anything from putting something on a scale to an observation about the phrasing of a given text. Likewise, theories may range from strict mathematical formula to less tangible things like "the intention of the author is [...]". The important part is, that the theory is testable by experiment, which in turn can range from building a physical experiment, over checking whether other works of an author support the theory about his intentions, to conducting a poll.

This is why I said "within the boundaries of its subject": A physicist will conduct physical experiments, testing whether nature does indeed follow the mathematical laws that the theorists came up with. A historian will search and read old documents/inscriptions to narrow in on what really happened. A researcher studying supernovae will rely on both random observations of supernovae (as many as possible) and on numeric simulations of what processes take place as a star ends its life. In each case, the words "theory" and "experiment" have radically different meanings, but they still play the same role in producing knowledge.

A "science" that lacks any form of scientific method (as outlined above), is simply not science, and does not produce any valuable results. It's guesswork at best, and utter, harmful bullshit at its worst.


tl;dr As a litmus-test, you can see how experts in a topic react when one of their field's well-established predictions is contradicted by experiment:

  • If they're scientific: They may be fascinated and eager to learn everything they can about what happened.

  • If they're pseudo-scientific: They may be disinterested, dismissive, defensive, etc..

Scientists seek to improve their understanding; things that disprove them are exciting!

Generally speaking, real scientists are interested in improving their ability to make predictions; so, anything that contradicts their expectations tends to be met with great excitement. Likewise, they tend to find experiments that prove them right to be relatively boring, even if they appreciate the value of confirmational-studies.

However, if someone gets defensive about being disproven, or is otherwise disinterested or dismissive of (verified, statistically significant) experimental contradictions, then they're probably not thinking like a scientist.

To be clear:

  1. It's about improving knowledge; experiments are more exciting/interesting when they provide more information. So:

    • Experiments that confirm things that scientists already firmly believed tend to be boring, because the scientist is left with the same beliefs afterwards.

    • Experiments that refute things that the scientists firmly disbelieved tend to be boring, because the scientist is left with the same beliefs afterwards.

    • Experiments that provide statistically weak contradictions can tend to be boring because the scientist can already explain them: they weren't statistically reliable.

  2. Scientists can come up with speculative theories, e.g. in Theoretical Physics. When such theories are thought to be possible-but-not-certain, then:

    • Scientists may be excited for either a confirmation or refutation. Because, either way, they've learned something.

    • Scientists will tend to be clear that such speculative theories are speculative; they won't present speculations as established fact.

    • For personal career-advancement reasons, etc., a flawed scientist might be somewhat defensive of a pet theory. This might reflect poorly on their objectivity to some degree or another, though it's generally not enough to broadly label them as a pseudo-scientist (in small doses).

    • A scientist might reasonably pursue a speculative theory that they perceive to be unlikely. So long as they're objective in assessing it – for example, they don't misrepresent it as being more/less likely than it is – exploring a speculative hypothesis isn't pseudo-scientific.

  3. Malpractices like faking results, misrepresenting experiments, etc., are red-flags for pseudo-science, even if an alleged-researcher is doing work in a normally respectable field.


In an adversarial system, no litmus test for pseudoscience can survive

From a zoomed out perspective, there can be no simple test, because this is an adversarial system.

Science is the study of predictive patterns in verifiable observations. Pseudoscience, broadly speaking, is anything that tries to look like science but isn't. Pseudoscience is an intentional creation, so the deception can benefit from its human creator's knowledge of their adversary's filters.

We can detect low quality fakes

Some knowledge of the scientific method can help crack low-quality fakes.

For example:

  • If something doesn't talk at all about the experiments that motivate its findings and jumps straight into conclusions, it's usually not science.
  • If something makes claims about experiments that its sources don't have the resources to run, it's probably not science. For example, an article about quantum physics from someone who doesn't have access to the expensive lab equipment needed to run quantum physical experiments is probably peddling pseudoscience.
  • If something extrapolates from unscientific priors (ghosts, chi, telepathy) it's probably not science (though still could be if it's got reproducible results to justify its conclusions).
  • If something makes claims about a scientific discipline that contradict verified models, it's probably not science. For example, something that claims human psychology is affected by gravity from distant stars is probably pseudoscience, because mathematically we're more affected by the gravity of passing airplanes. (Though it still could be science if it's got reproducible results to justify its conclusions)

We can rely on others to invest effort in debunking medium quality fakes for us

The purveyors of pseudoscience know about all of the above tests. Some of them don't care (because their target audience doesn't apply those tests), but some of them do. A company who is trying to sell an expensive dietary product to educated people would want to be able to make claims about its product's health benefits that will withstand cursory scrutiny, so they will fund pseudoscience that doesn't fall into those traps.

In some of these cases, we can wait to see if their findings get accepted by scientific journals. An attention grabbing pop-sci headline is sourced from a pre-print (or a low-quality pay-to-publish journal), it might be science, but also might not be. During peer review, other experts in the field will inspect their methods to verify the soundness of their experiment and judge whether the results of the experiment justify the claims made, which weeds out medium-quality fake science.

High quality fakes will slip through

Unfortunately, extremely well funded adversaries (such as nation states and billion-dollar corporations) can fund experts to write pseudoscience for them, so that it avoids the markings that would be recognized by other experts and gets through peer review. This is because science is a process, not a result, and it is impossible to know if the method was followed seeing only the artifacts produced.

P-Hacking follows all of the steps of the scientific method, but follows them out of order. Since all the steps are followed, it is possible for a P-Hacked paper to legitimately produce the artifacts that would have been produced if the same experiment was run scientifically and supported their eventual conclusion.

It can be impossible to tell a well-written P-Hacked paper from a scientific one.

Our last line of defense is reproducibility

Science's last line of defense is to repeat experiments. Real science will give the same results on reproduction (to a known margin of error, which depends on the discipline). High quality pseudoscientific claims will eventually disappear when someone runs the claimed experiment again.

For now, we can (cautiously) accept that any result which has been borne out across a multitude of experiments, observed in longitudinal studies, and that has been broadly accepted by the scientific community for at least a decade is probably scientific.

For everything else, there's skepticism.


I'm not sure this will meet the standards of the pipe-smoking philosophers out there but here's what I would consider the minimum standard for a claim to be 'scientific':

  • It is falsifiable
  • Attempts to falsify it are repeatable

Note: It need not be 'true'.

In formal 'hard' science, one forms a 'hypothesis' (not to be confused with a theory) and then attempts are made to disprove it. When enough attempts to disprove the hypothesis have failed, scientists start to accept that the hypothesis may be correct. This process cannot occur if a claims is unfalsifiable or cannot be repeatedly tested.

Pseudo-science, as I see it, works in the opposite direction. A theory is formed, typically based on a belief or idea but sometimes it can be based on some sort of evidence. Then it is either simply assumed to be true based on some sort of logic (it doesn't need to be valid logic) or if it is tested, the test is meant to find evidence that the claim is correct.

It's generally easy to find evidence of unfalsifiable claims and for falsifiable claims you can almost always point to something as evidence. For example, someone claims a homeopathic remedy cures the common cold. After using the remedy for a week, many people report their cold is gone. But without a control group, the claim that the remedy cures the common cold cannot be falsified. While the example claim here can be tested scientifically, claims that has been scientifically proven using unscientific methods clearly makes it 'psuedo-scientific'. It's an attempt to make it appearance like scientific knowledge without actual science.

Again, a claim doesn't need to be correct to be scientific and a psuedo-scientific claim doesn't need to be false. Moreover, the idea that only scientific claims should be considered to be real knowledge is flawed. The distinction is the process and how science can be used to create knowledge that we are highly confident in.

One thing that I think can be confusing is that there are many situations where we cannot actively create repeatable experiments and are forced to rely on 'natural experiments'. This is common in fields in like archaeology, astronomy, geology, and social sciences. But even in these scenarios, for a claim to be considered scientific, it must be falsifiable and the natural experiments must have some way that allows for repeated attempts at invalidation. An example is if a very unique and unusual fossil is found. It is interesting and could create new ideas but it's only scientific is disproving some other claim if an until more related evidence is found. These limits are inherent in some fields but the standards of what is a 'scientific claim' remains the same.

Lastly, in the past, I've encountered people saying that scientific claims are, by definition, claims made by a 'scientists' (as if that's some sort of official designation). This idea is way off. A lot of pseudo-scientific claims come from people we would consider professional scientists or people trained in science.


Pseudoscience is often built upon a scientific basis but the logic isn't sound.


  1. Science proves that gluten triggers inflammation in some people and has no essential nutrients.
  2. Bread is a major source of gluten.
  3. Therefore, gluten-free bread is a healthier alternative.


  1. Science proves that injecting diluted viruses into a system increases the body's immunity to viruses.
  2. Science proves that consuming small quantities of snake venom increases tolerance to venom.
  3. Therefore, giving small quantities of onion increases tolerance to hay fever.

If it can be proven to be false, it is no longer pseudoscience, but antiscience.

The nature of pseudoscience is that it is scientifically plausible, but takes too much effort to prove or disprove.


Yes, broadly there are two ways of testing what is science…

First, what studies have been published to support the idea you're challenging, and how scientific were they?

To go further, to what extent can you repeat the experiments reported, and see the same results?

Repeating the experiments might be expensive and what has that to do with the science?


Going a little bit outside philosophy domain and all those deep and knowledgeable answers already provided, here are some of the statements that always trigger my pseudoscience alarm:

  • Claiming to be 'holistic'
  • Having a nice website to advertise their therapy/product though you haven't heard about it before from actual professionals in the field
  • Having anecdotal evidence (e.g. singling out a patient or two who seemed to benefit from their pseudoscience therapy) in comparison to statistical evidence
  • Being angry about colleagues not accepting their product/therapy
  • Referring to tradition/religion, etc.
  • Being way overpriced in comparison to what you can get from 'mainstream' science

The concept of 'pseudoscience' traces back to Karl Popper's theoretical framework, which has largely been abandoned by the philosophy of science. It's importance is only historical; no one should really be using he term these days. Frankly, the concept only has any currency because it has become part of the ongoing political struggles between advocates for science and religion; there is a certain class of die-hard anti-metaphysicians in the non-philosophical community who use it to 'debunk' anything outside the scientific orthodoxy.

Pseudoscience as a concept is actually anti-scientific: it tries to establish what is and is not scientifically valid prior to any theoretical analysis or empirical research. The fact is that everything we currently consider 'established' science appeared at its beginning as nonsense running counter to accepted theory. In the 1970s, for instance, the idea that dinosaurs had anything in common with avian species was considered ridiculous; now it is a generally accepted truth. Sciences only became 'sciences' through a drawn-out process of argumentation, experimentation, and usage that may drag out over decades or centuries; one generation's pseudoscience might easily become the next generation's hard truth.

The categories you should be thinking about are:

  • Non-sciences: activités that don't seem or claim to be scientific at all.
  • Technologies: things that seem 'sciencey' and that may have practical use, but aren't studied deeply
  • Sciences: Things that are studied with some form of empirical rigor. This includes:
    • Failed science: theories that are empirically tested and just don't pan out
    • Successful science: theories that survive empirical testing

There are, of course, people who cling to failed sciences long past the expiration date, and other people who swear by things that are non-sciences or technologies. Welcome to the human race... But Popper's attempt the analytically distinguish between 'real' science and 'pseudo' science was more or less doomed from the get-go.

  • 3
    Ted -- this claim is untrue. Popper's demarcation is currently being deployed to great utility in challenging the nonsense of Multiverse and Simulation speculative "science".
    – Dcleve
    Commented Aug 16, 2022 at 1:29
  • @Dcleve: Used by whom? In what context? Popper's demarcation is analytically unsound; it can be used rhetorically (within some purely political context) but it cannot be used to objectively remark anything. I mean, I like the basic demarcation concept on an intuitive level, but the application is too subjective to have any real punch. Commented Aug 16, 2022 at 3:56
  • Ted -- no arguments satisfy "analytically sound", nor can ever be objective. LOGIC after all is pluralist. Your critiques are an impossible standard fallacy. Popper and Lakatos provide the best tools we have to sort out what is wrong with Susskind's Cosmic Landscape hypothesis, and irrefutable "but everything could be a simulation with features that are different from our simulations" speculations. Same with "God designed it but we can't know what God is like, so can't say that any X is undesigned because different from what a designer would do" b/s.
    – Dcleve
    Commented Aug 16, 2022 at 4:35
  • @Dcleve: No, that's mostly incorrect. Just as there are dimensions of rigor in scientific analysis, there are dimensions of rigor in philosophical work. The notion of pseudoscience has always been troubled by its inability to make clear distinctions. The concept of pseudoscience is unfalsifiable in and of itself; it ultimately boils down to a subjective assessment of 'distaste' for a particular subject matter. Commented Aug 16, 2022 at 19:56
  • Ted -- you are repeating the prior impossible standard. ALL philosophy, just like all of science, is judgement calls. No philosophy can satisfy "objectivity". Popper's approach, like all other philosophy, fails the analytic standard of certainty. As, of course, does analyticity itself. What Popper offers is an excellent pragmatically useful standard. Pragmatism is the only criteria that ultimately we can have, and yes, it is subjective, and yes finding pragmatic truth relies upon a consensus of honest people with good judgement.
    – Dcleve
    Commented Aug 16, 2022 at 21:41

I think most of these subjects are very complex, to call the entirety of such subjects science or pseudoscience would be wrong. I rather think that the science or pseudoscience tag can only be given to one particular argument of any subject. For example, astrology is pseudoscientific, yet some astrologers might agree that Earth revolves around the Sun, that's a scientific fact! You can't generalize a tag like that. Scientific analysis best works by dealing things piece by piece. Similarly many ayurvedic medicine work like eating honey and turmeric can cure cold, naturopathy has many scientific aspects, but there are not entirely backed up. Similarly for other subjects.

Biology on the contrary is definitely science! But some bio-reasearch claims without evidence backing can be called pseudoscientific sometimes!

One must remember that these are adjectives, and you cant give adjectives to subjects generally, like geology's boring, maths is fun etc.

  • A piece of information isn't science, science is a method of gathering data and validating or falsifying theories. It's not impossible for someone to be right on accident, but that's not what makes it scientific.
    – haxor789
    Commented Aug 18, 2022 at 10:13

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .